|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/llm-model-quantization-an-overview/
课程评论：没有评论
**Coursera LLM模型量化：概览课程总结** 本课程深入探讨了模型量化在大型语言模型（LLM）中的应用，旨在为机器学习、自然语言处理以及AI模型优化感兴趣的学生、专业人士和爱好者提供全面的理解。 **核心内容与目标：** * **基础原理与必要性：** 理解模型量化的基本概念，以及其在LLM领域的重要性、益处和驱动因素。 * **量化类型与方法：** 详细介绍并对比后训练量化（PTQ）、量化感知训练（QAT）和动态量化等主流量化技术。 * **主流框架应用：** 学习使用PyTorch、TensorFlow（含TensorFlow Lite）、ONNX以及NVIDIA TensorRT等关键框架进行模型量化。 * **量化模型评估：** 掌握如何通过准确率、延迟、吞吐量等性能指标，以及困惑度、BLEU、ROUGE等质量指标来评估量化模型，并了解相关的人工和自动评估技术。 * **量化模型部署：** 学习在边缘设备和云平台（如OpenAI和Azure OpenAI）上部署量化LLM的策略，并探讨部署中的权衡、优势与挑战。 **课程结构：** 课程共分为五个部分： 1. **模型量化导论：** 概述量化概念、重要性及基本原理。 2. **量化类型与方法：** 深入解析PTQ、QAT和动态量化，并进行比较。 3. **量化框架：** 介绍在PyTorch、TensorFlow、ONNX和NVIDIA TensorRT中量化的使用。 4. **量化模型评估：** 关注性能与质量指标，以及评估方法。 5. **量化模型部署：** 涵盖边缘与云端部署的策略和考量。 **目标受众：** 本课程适合AI和机器学习爱好者、数据科学家和工程师、计算机科学及相关领域学生，以及AI和NLP行业的专业人士。
Course Description:This course offers a deep dive into the world of model quantization, specifically focusing on its application in Large Language Models (LLMs). It is tailored for students, professionals, and enthusiasts interested in machine learning, natural language processing, and the optimization of AI models for various platforms. The course covers fundamental concepts, practical methodologies, various frameworks, and real-world applications, providing a well-rounded understanding of model quantization in LLMs.Course Objectives:Understand the basic principles and necessity of model quantization in LLMs.Explore different types and methods of model quantization, such as post-training quantization, quantization-aware training, and dynamic quantization.Gain proficiency in using major frameworks like PyTorch, TensorFlow, ONNX, and NVIDIA TensorRT for model quantization.Learn to evaluate the performance and quality of quantized models in real-world scenarios.Master the deployment of quantized LLMs on both edge devices and cloud platforms.Course Structure:Lecture 1: Introduction to Model QuantizationOverview of model quantizationSignificance in LLMsBasic concepts and benefitsLecture 2: Types and Methods of Model QuantizationPost-training quantizationQuantization-aware trainingDynamic quantizationComparative analysis of each typeLecture 3: Frameworks for Model QuantizationPyTorch's quantization toolsTensorFlow and TensorFlow LiteONNX quantization capabilitiesNVIDIA TensorRT's role in quantizationLecture 4: Evaluating Quantized ModelsPerformance metrics: accuracy, latency, and throughputQuality metrics: perplexity, BLEU, ROUGEHuman evaluation and auto-evaluation techniquesLecture 5: Deploying Quantized ModelsStrategies for edge device deploymentCloud platform deployment: OpenAI and Azure OpenAITrade-offs, benefits, and challenges in deploymentTarget Audience:AI and Machine Learning enthusiastsData Scientists and EngineersStudents in Computer Science and related fieldsProfessionals in AI and NLP industries
